Re: [fedora-java] Re: [jonas-team] Monolog on gcj

[Andrew Haley <aph@xxxxxxxxxx>]

Vadim Nasardinov writes:
 > On Friday 19 August 2005 11:20, Andrew Haley wrote:
 > >  > This responsibility should most certainly not be born by the
 > >  > LoggerFactory class.
 > > 
 > > I don't think that makes sense, because PropertiesConfAccess might
 > > well die before the clients call LoggerFactory.getLogger().
 > > 
 > > The method getLoggerFactory() calls PropertiesConfAccess.load(),
 > > which does
 > > 
 > >         new PropertiesConfAccess().read(prop, mf, mf, mf);
 > > 
 > > so, the instance of PropertiesConfAccess is dead even before the
 > > method LoggerFactory.getLogger() returns.
 > 
 > This doesn't change the fact that LoggerFactory is the wrong place to
 > intervene.

Yes it does.

 > All this says is that the PropertiesConfAccess (PCA) instance may
 > have to be kept undead.

That can't be done with the existing Monolog interface, as I
explained.  PropertiesConfAccess.load() does not return an instance of
PropertiesConfAccess.

 > Or, it says that the PCA, before dying a quiet death, has to return
 > the newly instantiated collection of loggers to whoever made the
 > call to the PCA so that they -- the caller -- can decide whether or
 > not they want to keep said loggers around.
 > 
 > Whatever the right patch turns out to be, the one we have on the table
 > right now is certainly not it.

So come up with something that works, rather than throwing about
wildly impractical suggestions.

 > That said, I don't really care if Monolog leaks memory.  If everyone
 > else is happy with the patch, then so be it.

Excellent.

Andrew.